├── .gitignore ├── LICENSE ├── README.md ├── basic_pipeline.py ├── dataloaders ├── __init__.py ├── algos.pyx ├── basic_collate.py ├── datasets_GPT.py ├── galatica_smiles_collator.py ├── gimlet_collator.py ├── gin_text_collator.py ├── gpt3_smiles_collator.py ├── graph_text_transform.py ├── graphormer_collator.py ├── graphormer_transform.py ├── kvplm_smiles_collator.py ├── molecule_datasets.py ├── momu_collator.py ├── splitters.py └── wrap_dataset.py ├── downstream_test.py ├── downstream_test.sh ├── fig ├── gimlet.png └── mol_tasks.png ├── finetune_property_prediction_graph_only.py ├── finetune_property_prediction_graph_only.sh ├── model ├── Arguments.py ├── GIMLET │ ├── GIMLETEncoderStack.py │ ├── GIMLETTransformerForConditionalGeneration.py │ ├── GIMLETTransformerForGraphOnly.py │ ├── __init__.py │ └── gnn_model.py ├── KVPLM │ ├── file_utils.py │ ├── kvplm_model.py │ ├── modeling.py │ ├── optimization.py │ ├── smtokenization.py │ └── tokenization.py ├── MoMu │ ├── .gitignore │ ├── KVPLMutils.py │ ├── MoMu_Model.py │ ├── README.md │ ├── __init__.py │ ├── data_provider │ │ ├── match_dataset.py │ │ └── sent_dataset.py │ ├── drug_discover.py │ ├── main.py │ ├── model │ │ ├── __init__.py │ │ ├── bert.py │ │ ├── contrastive_gin.py │ │ └── gin_model.py │ ├── optimization.py │ └── utils │ │ ├── GraphAug.py │ │ ├── lr.py │ │ └── text_translation_metrics.py ├── __init__.py ├── graphormer │ ├── models │ │ ├── __init__.py │ │ ├── graphormer.py │ │ ├── graphormer_3d.py │ │ └── graphormer_multitask.py │ └── modules │ │ ├── __init__.py │ │ ├── fairseq_dropout.py │ │ ├── graphormer_graph_encoder.py │ │ ├── graphormer_graph_encoder_layer.py │ │ ├── graphormer_layers.py │ │ ├── layer_drop.py │ │ ├── layer_norm.py │ │ ├── multihead_attention.py │ │ ├── quant_noise.py │ │ └── softmax.py └── llm.py ├── pretraining_gimlet.py ├── pretraining_graph_only.py ├── prompt_data ├── README.md ├── chembl_full_gen.py ├── chembl_full_gen.sh ├── chembl_gen.py ├── chembl_gen.sh └── transform.py ├── prompts ├── ablated_prompt_downstream_task.json ├── augment_prompt.py ├── augmented_prompt_downstream_task.json ├── augmented_prompt_pretrain.json ├── augmented_selected_prompt_downstream_task.json ├── generate_pcba_prompt.py ├── generate_pretrain_dataset.py ├── generate_pretrain_dataset_merge.py ├── generate_toxcast_prompt.py ├── pretrain_data_crawler.py ├── prompt_ablation.py ├── prompt_downstream_task.json ├── prompt_pretrain.json ├── select_prompt.py └── selected_prompt_downstream_task.json └── result_statistic.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/.gitignore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/README.md -------------------------------------------------------------------------------- /basic_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/basic_pipeline.py -------------------------------------------------------------------------------- /dataloaders/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/__init__.py -------------------------------------------------------------------------------- /dataloaders/algos.pyx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/algos.pyx -------------------------------------------------------------------------------- /dataloaders/basic_collate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/basic_collate.py -------------------------------------------------------------------------------- /dataloaders/datasets_GPT.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/datasets_GPT.py -------------------------------------------------------------------------------- /dataloaders/galatica_smiles_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/galatica_smiles_collator.py -------------------------------------------------------------------------------- /dataloaders/gimlet_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/gimlet_collator.py -------------------------------------------------------------------------------- /dataloaders/gin_text_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/gin_text_collator.py -------------------------------------------------------------------------------- /dataloaders/gpt3_smiles_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/gpt3_smiles_collator.py -------------------------------------------------------------------------------- /dataloaders/graph_text_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/graph_text_transform.py -------------------------------------------------------------------------------- /dataloaders/graphormer_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/graphormer_collator.py -------------------------------------------------------------------------------- /dataloaders/graphormer_transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/graphormer_transform.py -------------------------------------------------------------------------------- /dataloaders/kvplm_smiles_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/kvplm_smiles_collator.py -------------------------------------------------------------------------------- /dataloaders/molecule_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/molecule_datasets.py -------------------------------------------------------------------------------- /dataloaders/momu_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/momu_collator.py -------------------------------------------------------------------------------- /dataloaders/splitters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/splitters.py -------------------------------------------------------------------------------- /dataloaders/wrap_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/dataloaders/wrap_dataset.py -------------------------------------------------------------------------------- /downstream_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/downstream_test.py -------------------------------------------------------------------------------- /downstream_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/downstream_test.sh -------------------------------------------------------------------------------- /fig/gimlet.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/fig/gimlet.png -------------------------------------------------------------------------------- /fig/mol_tasks.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/fig/mol_tasks.png -------------------------------------------------------------------------------- /finetune_property_prediction_graph_only.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/finetune_property_prediction_graph_only.py -------------------------------------------------------------------------------- /finetune_property_prediction_graph_only.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/finetune_property_prediction_graph_only.sh -------------------------------------------------------------------------------- /model/Arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/Arguments.py -------------------------------------------------------------------------------- /model/GIMLET/GIMLETEncoderStack.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/GIMLET/GIMLETEncoderStack.py -------------------------------------------------------------------------------- /model/GIMLET/GIMLETTransformerForConditionalGeneration.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/GIMLET/GIMLETTransformerForConditionalGeneration.py -------------------------------------------------------------------------------- /model/GIMLET/GIMLETTransformerForGraphOnly.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/GIMLET/GIMLETTransformerForGraphOnly.py -------------------------------------------------------------------------------- /model/GIMLET/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /model/GIMLET/gnn_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/GIMLET/gnn_model.py -------------------------------------------------------------------------------- /model/KVPLM/file_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/file_utils.py -------------------------------------------------------------------------------- /model/KVPLM/kvplm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/kvplm_model.py -------------------------------------------------------------------------------- /model/KVPLM/modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/modeling.py -------------------------------------------------------------------------------- /model/KVPLM/optimization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/optimization.py -------------------------------------------------------------------------------- /model/KVPLM/smtokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/smtokenization.py -------------------------------------------------------------------------------- /model/KVPLM/tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/KVPLM/tokenization.py -------------------------------------------------------------------------------- /model/MoMu/.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/.gitignore -------------------------------------------------------------------------------- /model/MoMu/KVPLMutils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/KVPLMutils.py -------------------------------------------------------------------------------- /model/MoMu/MoMu_Model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/MoMu_Model.py -------------------------------------------------------------------------------- /model/MoMu/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/README.md -------------------------------------------------------------------------------- /model/MoMu/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /model/MoMu/data_provider/match_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/data_provider/match_dataset.py -------------------------------------------------------------------------------- /model/MoMu/data_provider/sent_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/data_provider/sent_dataset.py -------------------------------------------------------------------------------- /model/MoMu/drug_discover.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/drug_discover.py -------------------------------------------------------------------------------- /model/MoMu/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/main.py -------------------------------------------------------------------------------- /model/MoMu/model/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /model/MoMu/model/bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/model/bert.py -------------------------------------------------------------------------------- /model/MoMu/model/contrastive_gin.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/model/contrastive_gin.py -------------------------------------------------------------------------------- /model/MoMu/model/gin_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/model/gin_model.py -------------------------------------------------------------------------------- /model/MoMu/optimization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/optimization.py -------------------------------------------------------------------------------- /model/MoMu/utils/GraphAug.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/utils/GraphAug.py -------------------------------------------------------------------------------- /model/MoMu/utils/lr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/utils/lr.py -------------------------------------------------------------------------------- /model/MoMu/utils/text_translation_metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/MoMu/utils/text_translation_metrics.py -------------------------------------------------------------------------------- /model/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/__init__.py -------------------------------------------------------------------------------- /model/graphormer/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/models/__init__.py -------------------------------------------------------------------------------- /model/graphormer/models/graphormer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/models/graphormer.py -------------------------------------------------------------------------------- /model/graphormer/models/graphormer_3d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/models/graphormer_3d.py -------------------------------------------------------------------------------- /model/graphormer/models/graphormer_multitask.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/models/graphormer_multitask.py -------------------------------------------------------------------------------- /model/graphormer/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/__init__.py -------------------------------------------------------------------------------- /model/graphormer/modules/fairseq_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/fairseq_dropout.py -------------------------------------------------------------------------------- /model/graphormer/modules/graphormer_graph_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/graphormer_graph_encoder.py -------------------------------------------------------------------------------- /model/graphormer/modules/graphormer_graph_encoder_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/graphormer_graph_encoder_layer.py -------------------------------------------------------------------------------- /model/graphormer/modules/graphormer_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/graphormer_layers.py -------------------------------------------------------------------------------- /model/graphormer/modules/layer_drop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/layer_drop.py -------------------------------------------------------------------------------- /model/graphormer/modules/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/layer_norm.py -------------------------------------------------------------------------------- /model/graphormer/modules/multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/multihead_attention.py -------------------------------------------------------------------------------- /model/graphormer/modules/quant_noise.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/quant_noise.py -------------------------------------------------------------------------------- /model/graphormer/modules/softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/graphormer/modules/softmax.py -------------------------------------------------------------------------------- /model/llm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/model/llm.py -------------------------------------------------------------------------------- /pretraining_gimlet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/pretraining_gimlet.py -------------------------------------------------------------------------------- /pretraining_graph_only.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/pretraining_graph_only.py -------------------------------------------------------------------------------- /prompt_data/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/README.md -------------------------------------------------------------------------------- /prompt_data/chembl_full_gen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/chembl_full_gen.py -------------------------------------------------------------------------------- /prompt_data/chembl_full_gen.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/chembl_full_gen.sh -------------------------------------------------------------------------------- /prompt_data/chembl_gen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/chembl_gen.py -------------------------------------------------------------------------------- /prompt_data/chembl_gen.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/chembl_gen.sh -------------------------------------------------------------------------------- /prompt_data/transform.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompt_data/transform.py -------------------------------------------------------------------------------- /prompts/ablated_prompt_downstream_task.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/ablated_prompt_downstream_task.json -------------------------------------------------------------------------------- /prompts/augment_prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/augment_prompt.py -------------------------------------------------------------------------------- /prompts/augmented_prompt_downstream_task.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/augmented_prompt_downstream_task.json -------------------------------------------------------------------------------- /prompts/augmented_prompt_pretrain.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/augmented_prompt_pretrain.json -------------------------------------------------------------------------------- /prompts/augmented_selected_prompt_downstream_task.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/augmented_selected_prompt_downstream_task.json -------------------------------------------------------------------------------- /prompts/generate_pcba_prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/generate_pcba_prompt.py -------------------------------------------------------------------------------- /prompts/generate_pretrain_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/generate_pretrain_dataset.py -------------------------------------------------------------------------------- /prompts/generate_pretrain_dataset_merge.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/generate_pretrain_dataset_merge.py -------------------------------------------------------------------------------- /prompts/generate_toxcast_prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/generate_toxcast_prompt.py -------------------------------------------------------------------------------- /prompts/pretrain_data_crawler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/pretrain_data_crawler.py -------------------------------------------------------------------------------- /prompts/prompt_ablation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/prompt_ablation.py -------------------------------------------------------------------------------- /prompts/prompt_downstream_task.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/prompt_downstream_task.json -------------------------------------------------------------------------------- /prompts/prompt_pretrain.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/prompt_pretrain.json -------------------------------------------------------------------------------- /prompts/select_prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/select_prompt.py -------------------------------------------------------------------------------- /prompts/selected_prompt_downstream_task.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/prompts/selected_prompt_downstream_task.json -------------------------------------------------------------------------------- /result_statistic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/zhao-ht/GIMLET/HEAD/result_statistic.py --------------------------------------------------------------------------------